Calculatrice d’erreur relative
The detailed guide below is currently available in English.
The percent error formula
Percent error measures how far a measurement sits from an accepted reference, scaled by that reference: percent error = |observed − true| ÷ |true| × 100. Measuring 8.6 against a true value of 10 gives an error of |−1.4| ÷ 10 × 100 = 14%. The absolute value signs keep the size of the mistake positive; whether you were high or low is reported separately, and many lab reports ask for exactly that pairing. Taking the absolute value of the true value in the denominator also makes the formula well-defined when the reference itself is negative, such as a temperature below zero.
Questions fréquentes
Can percent error be negative?
The size of the error cannot — the formula takes the absolute value, so results are zero or positive. The direction of the miss (observed above or below true) is shown separately so nothing is hidden.
What counts as an acceptable percent error?
It depends entirely on the field. A chemistry titration within 1–2% is solid work, while a rough weather forecast within 10% may be excellent. Compare your result against the precision your task actually requires.